Investigating Unconfirmed AIDS Claims: Aids Never Isolated or Photographed?

[adrenaline]

The ELISA detects antibodies to HIV +/- p24 Antigen (in the case of 4th generation assays). It is a highly sensitive and highly specific assay, but is capable of giving significant numbers of false positives in a low prevalence population (where the pre-test odds are low). This is a common pitfall : low prevalence compromises positive predictive value in any test with less than 100% specificity.

The Western Blot (or immunoblot) detects specific antibodies to weight sorted HIV-1 proteins belonging to the 3 classes (gag, pol, env). It is a less sensitive but more specific assay than the ELISA, and is often used for confirmation and disease follow-up. The Western Blot pattern can be positive, negative or indeterminate and there are different criteria for reading them (CDC/ASTPHLD, American Red Cross, CRSS etc.) The pattern can evolve with disease progression. HIV-2 specific WBs are available.

Nucleic acid tests are the most sensitive, most specific and most expensive. They are difficult to perform with expertise (if done improperly, false negatives due to sample matrix inhibition and false positives due to sample-sample carryover contamination abound). They are uncommonly used in routine clinical diagnosis and are generally reserved for screening blood products prior to human transfusion. They are also sometimes useful in testing neonates with vertically acquired maternal antibodies to HIV. You can test for viral RNA or proviral DNA (which is the form that integrates into human cells). You can use different assay methodologies (PCR for DNA, RT-PCR for RNA and bDNA (branched chain DNA assay)).

very true which is why no medical test should be ordered unless the pretest probability for the disease is high and the local prevalence of the disease is taken into consideration. Ordering a HIV test in a virginal nun is probably not a good idea. ( ELisa with reflex western blot if positive). For that reason I don't order CA 125s at all for ovarian cancer screening, the sensitivey and specificity suck. However, compared to any singular medical diagnostic test I have yet to find any medical diagnostic test ( both serological as well as say, radiographic ( ex helical cat scan for pumonary embolism where senstiviites and specificities range from 57-95% and 78%-98% respectively depending on the study and type of cat scan) that comes close to the HIV test! There is no doubt under the right circumstance it can produce false positives and negatives but there is no other medical diagnostic test that I can recall offhand that has such a good specifcity and sensitvity ( Thus the potential to diagnose a blood clot in the lungs is fraught with much more false negatives and positives.) (I was sleep deprived yesterday so I meant to say DNA isolation vs amplification). Thats why this guy bringing up the rate of false positives and false negatives as the main thrust of his argument is way off base. He has no idea that almost all other medical diagnostic testing are far less sensitive and specific!

 

[Curious3141]

(I was sleep deprived yesterday so I meant to say DNA isolation vs amplification).

The Western blot is an immunologic (antibody) test. It is not a nucleic acid (DNA/RNA) assay at all.

I agree with the rest of the stuff about sens/spec. etc.

 

[adrenaline]

The Western blot is an immunologic (antibody) test. It is not a nucleic acid (DNA/RNA) assay at all.

I agree with the rest of the stuff about sens/spec. etc.

you are right, i keep thinking of pcr and my lack of sleep is making me3 punchy.